Consulting project for my MS degree in Statistics from Texas A&M University
Beef and chicken are two of the most common conventional protein sources consumed
by Americans. In recent years, the United States has experienced escalating beef prices along with growing popularity of chicken. These trends have been particularly acute since the beginning of the global COVID-19 pandemic in early 2020, and the overlapping period of high inflation that continues to the present. Therefore, forecasting these trends would be valuable for consumers, as well as the food and restaurant industries. Furthermore, since these price trends reflect consumption trends, they may also help inform public policy regarding implications for public health and the environment.
This investigation compared the average prices of U.S. beef and chicken for 459 months from January 1, 1984 to March 1, 2022. Predictive models were developed for the time series data using classical and machine learning methods. A valid forecasting model with the with lowest error rate (MAPE) was selected for each data series. Prices were forecasted one year into the future on March 1, 2023.
On March 1, 2022, in the United States, the average price for 100% ground beef was
$4.76 per pound, and the average price for fresh, whole chicken was $1.72 per pound, for a
price difference of $3.04 per pound. Using the forecasting model that I selected for each data series, I made the following projections for March 1, 2023, with a 95% level of confidence. The average price of 100% ground beef is projected to be between $3.61 and $4.77 per pound. The average price of fresh, whole chicken is projected to be between $1.46 and $1.78 per pound. Therefore, the average price difference between 100% ground beef and fresh, whole chicken is projected to be between $2.14 and $2.99 per pound.
Statistical computing for data analysis was performed using R Markdown
MS in Statistics